Yixiong Liang

Yixiong Liang, Shuo Feng, Qing Liu et al.

Cervical abnormal cell detection is a challenging task as the morphological discrepancies between abnormal and normal cells are usually subtle. To determine whether a cervical cell is normal or abnormal, cytopathologists always take surrounding cells as references to identify its abnormality. To mimic these behaviors, we propose to explore contextual relationships to boost the performance of cervical abnormal cell detection. Specifically, both contextual relationships between cells and cell-to-global images are exploited to enhance features of each region of interest (RoI) proposals. Accordingly, two modules, dubbed as RoI-relationship attention module (RRAM) and global RoI attention module (GRAM), are developed and their combination strategies are also investigated. We establish a strong baseline by using Double-Head Faster R-CNN with feature pyramid network (FPN) and integrate our RRAM and GRAM into it to validate the effectiveness of the proposed modules. Experiments conducted on a large cervical cell detection dataset reveal that the introduction of RRAM and GRAM both achieves better average precision (AP) than the baseline methods. Moreover, when cascading RRAM and GRAM, our method outperforms the state-of-the-art (SOTA) methods. Furthermore, we also show the proposed feature enhancing scheme can facilitate both image-level and smear-level classification. The code and trained models are publicly available at https://github.com/CVIU-CSU/CR4CACD.

Shichao Kan, Yixiong Liang, Min Li et al.

A fundamental challenge in deep metric learning is the generalization capability of the feature embedding network model since the embedding network learned on training classes need to be evaluated on new test classes. To address this challenge, in this paper, we introduce a new method called coded residual transform (CRT) for deep metric learning to significantly improve its generalization capability. Specifically, we learn a set of diversified prototype features, project the feature map onto each prototype, and then encode its features using their projection residuals weighted by their correlation coefficients with each prototype. The proposed CRT method has the following two unique characteristics. First, it represents and encodes the feature map from a set of complimentary perspectives based on projections onto diversified prototypes. Second, unlike existing transformer-based feature representation approaches which encode the original values of features based on global correlation analysis, the proposed coded residual transform encodes the relative differences between the original features and their projected prototypes. Embedding space density and spectral decay analysis show that this multi-perspective projection onto diversified prototypes and coded residual representation are able to achieve significantly improved generalization capability in metric learning. Finally, to further enhance the generalization performance, we propose to enforce the consistency on their feature similarity matrices between coded residual transforms with different sizes of projection prototypes and embedding dimensions. Our extensive experimental results and ablation studies demonstrate that the proposed CRT method outperform the state-of-the-art deep metric learning methods by large margins and improving upon the current best method by up to 4.28% on the CUB dataset.

Jialong Huang, Gaojie Li, Shichao Kan et al.

Current cervical cytopathology whole slide image (WSI) screening primarily relies on detection-based approaches, which are limited in performance due to the expense and time-consuming annotation process. Multiple Instance Learning (MIL), a weakly supervised approach that relies solely on bag-level labels, can effectively alleviate these challenges. Nonetheless, MIL commonly employs frozen pretrained models or self-supervised learning for feature extraction, which suffers from low efficacy or inefficiency. In this paper, we propose an efficient framework for cervical cytopathology WSI classification using only WSI-level labels through unsupervised and weakly supervised learning. Given the sparse and dispersed nature of abnormal cells within cytopathological WSIs, we propose a strategy that leverages the pretrained foundation model to filter the top$k$ high-risk patches. Subsequently, we suggest parameter-efficient fine-tuning (PEFT) of a large foundation model using contrastive learning on the filtered patches to enhance its representation ability for task-specific signals. By training only the added linear adapters, we enhance the learning of patch-level features with substantially reduced time and memory consumption. Experiments conducted on the CSD and FNAC 2019 datasets demonstrate that the proposed method enhances the performance of various MIL methods and achieves state-of-the-art (SOTA) performance. The code and trained models are publicly available at https://github.com/CVIU-CSU/TCT-InfoNCE.

Jing Jin, Xu Liu, Te Gao et al.

Whole Slide Image (WSI) representation is critical for cancer subtyping, cancer recognition and mutation prediction.Training an end-to-end WSI representation model poses significant challenges, as a standard gigapixel slide can contain tens of thousands of image tiles, making it difficult to compute gradients of all tiles in a single mini-batch due to current GPU limitations. To address this challenge, we propose a method of dynamic residual encoding with slide-level contrastive learning (DRE-SLCL) for end-to-end WSI representation. Our approach utilizes a memory bank to store the features of tiles across all WSIs in the dataset. During training, a mini-batch usually contains multiple WSIs. For each WSI in the batch, a subset of tiles is randomly sampled and their features are computed using a tile encoder. Then, additional tile features from the same WSI are selected from the memory bank. The representation of each individual WSI is generated using a residual encoding technique that incorporates both the sampled features and those retrieved from the memory bank. Finally, the slide-level contrastive loss is computed based on the representations and histopathology reports ofthe WSIs within the mini-batch. Experiments conducted over cancer subtyping, cancer recognition, and mutation prediction tasks proved the effectiveness of the proposed DRE-SLCL method.

Ziyuan Ding, Yixiong Liang, Shichao Kan et al.

High resolution is crucial for precise segmentation in fundus images, yet handling high-resolution inputs incurs considerable GPU memory costs, with diminishing performance gains as overhead increases. To address this issue while tackling the challenge of segmenting tiny objects, recent studies have explored local-global fusion methods. These methods preserve fine details using local regions and capture long-range context information from downscaled global images. However, the necessity of multiple forward passes inevitably incurs significant computational overhead, adversely affecting inference speed. In this paper, we propose HRDecoder, a simple High-Resolution Decoder network for fundus lesion segmentation. It integrates a high-resolution representation learning module to capture fine-grained local features and a high-resolution fusion module to fuse multi-scale predictions. Our method effectively improves the overall segmentation accuracy of fundus lesions while consuming reasonable memory and computational overhead, and maintaining satisfying inference speed. Experimental results on the IDRiD and DDR datasets demonstrate the effectiveness of our method. Code is available at https://github.com/CVIU-CSU/HRDecoder.

Haojie Zhang, Yixiong Liang, Hulin Kuang et al.

Multimodal Biomedical Image Incremental Learning (MBIIL) is essential for handling diverse tasks and modalities in the biomedical domain, as training separate models for each modality or task significantly increases inference costs. Existing incremental learning methods focus on task expansion within a single modality, whereas MBIIL seeks to train a unified model incrementally across modalities. The MBIIL faces two challenges: I) How to preserve previously learned knowledge during incremental updates? II) How to effectively leverage knowledge acquired from existing modalities to support new modalities? To address these challenges, we propose MSLoRA-CR, a method that fine-tunes Modality-Specific LoRA modules while incorporating Contrastive Regularization to enhance intra-modality knowledge sharing and promote inter-modality knowledge differentiation. Our approach builds upon a large vision-language model (LVLM), keeping the pretrained model frozen while incrementally adapting new LoRA modules for each modality or task. Experiments on the incremental learning of biomedical images demonstrate that MSLoRA-CR outperforms both the state-of-the-art (SOTA) approach of training separate models for each modality and the general incremental learning method (incrementally fine-tuning LoRA). Specifically, MSLoRA-CR achieves a 1.88% improvement in overall performance compared to unconstrained incremental learning methods while maintaining computational efficiency. Our code is publicly available at https://github.com/VentusAislant/MSLoRA_CR.

Qing Liu, Haotian Liu, Wei Ke et al.

Feature reassembly is an essential component in modern CNN-based segmentation approaches, which includes feature downsampling and upsampling operators. Existing operators reassemble multiple features from a small predefined region into one for each target location independently. This may result in loss of spatial information, which could vanish activations caused by tiny lesions particularly when they cluster together. In this paper, we propose a many-to-many reassembly of features (M2MRF). It reassembles features in a dimension-reduced feature space and simultaneously aggregates multiple features inside a large predefined region into multiple target features. In this way, long range spatial dependencies are captured to maintain activations on tiny lesions. Experimental results on two lesion segmentation benchmarks, i.e. DDR and IDRiD, show that (1) our M2MRF outperforms existing feature reassembly operators; (2) equipped with our M2MRF, the HRNetv2 is able to achieve significant better performance to CNN-based segmentation methods and competitive even better performance to two recent transformer-based segmentation methods. Our code is made publicly available at https://github.com/CVIU-CSU/M2MRF-Lesion-Segmentation.

Yixiong Liang, Yuan Mao, Zhihong Tang et al.

In this paper we propose a very efficient method to fuse the unregistered multi-focus microscopical images based on the speed-up robust features (SURF). Our method follows the pipeline of first registration and then fusion. However, instead of treating the registration and fusion as two completely independent stage, we propose to reuse the determinant of the approximate Hessian generated in SURF detection stage as the corresponding salient response for the final image fusion, thus it enables nearly cost-free saliency map generation. In addition, due to the adoption of SURF scale space representation, our method can generate scale-invariant saliency map which is desired for scale-invariant image fusion. We present an extensive evaluation on the dataset consisting of several groups of unregistered multi-focus 4K ultra HD microscopic images with size of 4112 x 3008. Compared with the state-of-the-art multi-focus image fusion methods, our method is much faster and achieve better results in the visual performance. Our method provides a flexible and efficient way to integrate complementary and redundant information from multiple multi-focus ultra HD unregistered images into a fused image that contains better description than any of the individual input images. Code is available at https://github.com/yiqingmy/JointRF.

Yixiong Liang, Yuan Mao, Jiazhi Xia et al.

In this paper, we present a fast yet effective method for pixel-level scale-invariant image fusion in spatial domain based on the scale-space theory. Specifically, we propose a scale-invariant structure saliency selection scheme based on the difference-of-Gaussian (DoG) pyramid of images to build the weights or activity map. Due to the scale-invariant structure saliency selection, our method can keep both details of small size objects and the integrity information of large size objects in images. In addition, our method is very efficient since there are no complex operation involved and easy to be implemented and therefore can be used for fast high resolution images fusion. Experimental results demonstrate the proposed method yields competitive or even better results comparing to state-of-the-art image fusion methods both in terms of visual quality and objective evaluation metrics. Furthermore, the proposed method is very fast and can be used to fuse the high resolution images in real-time. Code is available at https://github.com/yiqingmy/Fusion.

Yixiong Liang, Zhihong Tang, Meng Yan et al.

Automated detection of cervical cancer cells or cell clumps has the potential to significantly reduce error rate and increase productivity in cervical cancer screening. However, most traditional methods rely on the success of accurate cell segmentation and discriminative hand-crafted features extraction. Recently there are emerging deep learning-based methods which train convolutional neural networks (CNN) to classify image patches, but they are computationally expensive. In this paper we propose an efficient CNN-based object detection methods for cervical cancer cells/clumps detection. Specifically, we utilize the state-of-the-art two-stage object detection method, the Faster-RCNN with Feature Pyramid Network (FPN) as the baseline and propose a novel comparison detector to deal with the limited data problem. The key idea is that classify the proposals by comparing with the reference samples of each category in object detection. In addition, we propose to learn the reference samples of the background from data instead of manually choosing them by some heuristic rules. Experimental results show that the proposed Comparison Detector yields significant improvement on the small dataset, achieving a mean Average Precision (mAP) of 26.3% and an Average Recall (AR) of 35.7%, both improving about 20 points compared to the baseline. Moreover, Comparison Detector improved AR by 4.6 points and achieved marginally better performance in terms of mAP compared with baseline model when training on the medium dataset. Our method is promising for the development of automation-assisted cervical cancer screening systems. Code is available at https://github.com/kuku-sichuan/ComparisonDetector.

Shichao Kan, Xuyang Zhang, Haojie Zhang et al.

Evaluating image captions without references remains challenging because global embedding similarity often misses fine-grained mismatches such as hallucinated objects, missing attributes, or incorrect relations. We propose MSD-Score, a reference-free metric that models image patch and text token embeddings as von Mises-Fisher mixtures on the unit hypersphere. Instead of treating each modality as a single point, MSD-Score formulates image-text matching as a multi-scale distributional scoring problem. Semantic discrepancies are quantified via a weighted bi-directional KL divergence and combined with global similarity in a multi-scale framework for both single- and multi-candidate evaluations. Extensive experiments show that MSD-Score achieves state-of-the-art correlation with human judgments among reference-free metrics. Beyond accuracy, its probabilistic formulation yields transparent and decomposable diagnostics of local grounding errors, providing a deterministic complementary signal to holistic similarity metrics and judge-based evaluators.

Shichao Kan, Yuhai Deng, Jiale Fu et al.

Retrieval-augmented generation (RAG) with large language models (LLMs) plays a crucial role in question answering, as LLMs possess limited knowledge and are not updated with continuously growing information. Most recent work on RAG has focused primarily on text-based or large-image retrieval, which constrains the broader application of RAG models. We recognize that object-level retrieval is essential for addressing questions that extend beyond image content. To tackle this issue, we propose a task of object retrieval for visual question answering with outside knowledge (OR-OK-VQA), aimed to extend image-based content understanding in conjunction with LLMs. A key challenge in this task is retrieving diverse objects-related images that contribute to answering the questions. To enable accurate and robust general object retrieval, it is necessary to learn embeddings for local objects. This paper introduces a novel unsupervised deep feature embedding technique called multi-scale group collaborative embedding learning (MS-GCEL), developed to learn embeddings for long-tailed objects at different scales. Additionally, we establish an OK-VQA evaluation benchmark using images from the BelgaLogos, Visual Genome, and LVIS datasets. Prior to the OK-VQA evaluation, we construct a benchmark of challenges utilizing objects extracted from the COCO 2017 and VOC 2007 datasets to support the training and evaluation of general object retrieval models. Our evaluations on both general object retrieval and OK-VQA demonstrate the effectiveness of the proposed approach. The code and dataset will be publicly released for future research.

Qing Liu, Haotian Liu, Yixiong Liang

Automated segmentation of hard exudates in colour fundus images is a challenge task due to issues of extreme class imbalance and enormous size variation. This paper aims to tackle these issues and proposes a dual-branch network with dual-sampling modulated Dice loss. It consists of two branches: large hard exudate biased learning branch and small hard exudate biased learning branch. Both of them are responsible for their own duty separately. Furthermore, we propose a dual-sampling modulated Dice loss for the training such that our proposed dual-branch network is able to segment hard exudates in different sizes. In detail, for the first branch, we use a uniform sampler to sample pixels from predicted segmentation mask for Dice loss calculation, which leads to this branch naturally be biased in favour of large hard exudates as Dice loss generates larger cost on misidentification of large hard exudates than small hard exudates. For the second branch, we use a re-balanced sampler to oversample hard exudate pixels and undersample background pixels for loss calculation. In this way, cost on misidentification of small hard exudates is enlarged, which enforces the parameters in the second branch fit small hard exudates well. Considering that large hard exudates are much easier to be correctly identified than small hard exudates, we propose an easy-to-difficult learning strategy by adaptively modulating the losses of two branches. We evaluate our proposed method on two public datasets and results demonstrate that ours achieves state-of-the-art performances.

Yao Xiang, Wanxin Sun, Changli Pan et al.

While most previous automation-assisted reading methods can improve efficiency, their performance often relies on the success of accurate cell segmentation and hand-craft feature extraction. This paper presents an efficient and totally segmentation-free method for automated cervical cell screening that utilizes modern object detector to directly detect cervical cells or clumps, without the design of specific hand-crafted feature. Specifically, we use the state-of-the-art CNN-based object detection methods, YOLOv3, as our baseline model. In order to improve the classification performance of hard examples which are four highly similar categories, we cascade an additional task-specific classifier. We also investigate the presence of unreliable annotations and cope with them by smoothing the distribution of noisy labels. We comprehensively evaluate our methods on test set which is consisted of 1,014 annotated cervical cell images with size of 4000*3000 and complex cellular situation corresponding to 10 categories. Our model achieves 97.5% sensitivity (Sens) and 67.8% specificity (Spec) on cervical cell image-level screening. Moreover, we obtain a mean Average Precision (mAP) of 63.4% on cervical cell-level diagnosis, and improve the Average Precision (AP) of hard examples which are valuable but difficult to distinguish. Our automation-assisted cervical cell reading method not only achieves cervical cell image-level classification but also provides more detailed location and category information of abnormal cells. The results indicate feasible performance of our method, together with the efficiency and robustness, providing a new idea for future development of computer-assisted reading system in clinical cervical screening.

Qing Liu, Beiji Zou, Yang Zhao et al.

Segmentation of optic disc (OD) and optic cup (OC) is critical in automated fundus image analysis system. Existing state-of-the-arts focus on designing deep neural networks with one or multiple dense prediction branches. Such kind of designs ignore connections among prediction branches and their learning capacity is limited. To build connections among prediction branches, this paper introduces gradient boosting framework to deep classification model and proposes a gradient boosting network called BoostNet. Specifically, deformable side-output unit and aggregation unit with deep supervisions are proposed to learn base functions and expansion coefficients in gradient boosting framework. By stacking aggregation units in a deep-to-shallow manner, models' performances are gradually boosted along deep to shallow stages. BoostNet achieves superior results to existing deep OD and OC segmentation networks on the public dataset ORIGA.

Yukun Zhou, Zailiang Chen, Hailan Shen et al.

The research on recognizing the most discriminative regions provides referential information for weakly supervised object localization with only image-level annotations. However, the most discriminative regions usually conceal the other parts of the object, thereby impeding entire object recognition and localization. To tackle this problem, the Dual-attention Focused Module (DFM) is proposed to enhance object localization performance. Specifically, we present a dual attention module for information fusion, consisting of a position branch and a channel one. In each branch, the input feature map is deduced into an enhancement map and a mask map, thereby highlighting the most discriminative parts or hiding them. For the position mask map, we introduce a focused matrix to enhance it, which utilizes the principle that the pixels of an object are continuous. Between these two branches, the enhancement map is integrated with the mask map, aiming at partially compensating the lost information and diversifies the features. With the dual-attention module and focused matrix, the entire object region could be precisely recognized with implicit information. We demonstrate outperforming results of DFM in experiments. In particular, DFM achieves state-of-the-art performance in localization accuracy in ILSVRC 2016 and CUB-200-2011.

Rui Kang, Yixiong Liang, Chunyan Lian et al.

The urine sediment analysis of particles in microscopic images can assist physicians in evaluating patients with renal and urinary tract diseases. Manual urine sediment examination is labor-intensive, subjective and time-consuming, and the traditional automatic algorithms often extract the hand-crafted features for recognition. Instead of using the hand-crafted features, in this paper, we exploit CNN to learn features in an end-to-end manner to recognize the urine particles. We treat the urine particles recognition as object detection and exploit two state-of-the-art CNN-based object detection methods, Faster R-CNN and SSD, as well as their variants for urine particles recognition. We further investigate different factors involving these CNN-based object detection methods for urine particles recognition. We comprehensively evaluate these methods on a dataset consisting of 5,376 annotated images corresponding to 7 categories of urine particles, i.e., erythrocyte, leukocyte, epithelial cell, crystal, cast, mycete, epithelial nuclei, and obtain a best mAP (mean average precision) of 84.1% while taking only 72 ms per image on a NVIDIA Titan X GPU.